公开(公告)号：DE69920893T2

公开(公告)日：2006-03-09

申请号：DE69920893

申请日：1999-07-13

Applicant: IBM

Inventor： FICHOU ALINE ,  LE PENNEC JEAN-FRANCOIS ,  GALAND CLAUDE

IPC: H04L12/54 ,  H04L47/762 ,  H04Q11/04

Abstract: Disclosed herein is a connection bandwidth management process and system for use in a high speed packet switching network. The network comprises a plurality of switching nodes interconnected through a plurality of communication links. Each of the switching nodes comprises means for switching packets from at least one input link to at least one output link. Each of the output links are coupled to at least one buffer in the switching node for queuing packets before they are transmitted over the output link. Each of the communication links supports the traffic of a plurality of user connections statistically multiplexed over the link. Each user connection is allocated an initial agreed-upon bandwidth through the network, with each of the communication links being possibly oversubscribed. The connection bandwidth management process according to the invention comprises the steps of: Link monitoring data on the communication links are periodically received in a network monitoring center, and stored in a computer memory in the network monitoring center. Then, one monitored link is selected and the corresponding link monitoring data are retrieved from the computer memory. The link monitored data retrieved for the selected link is analyzed, and it is determined whether the selected link is oversubscribed or not. If it is determined that the selected link is oversubscribed and that the link monitoring data for the selected link satisfies at least one predetermined condition the bandwidth initially allocated to each of the connections boarded on the selected link is reallocated, such that, the sum of the reallocated bandwidth of the connections boarded on the selected link is less or equal to the total bandwidth capacity of the selected link. The process recycles until all the monitored links have been selected.

公开(公告)号：DE3851887T2

公开(公告)日：1995-04-20

申请号：DE3851887

申请日：1988-07-18

Applicant: IBM

Inventor： ROSSO MICHELE ,  GALAND CLAUDE

IPC: G06F3/16 ,  G10L19/06 ,  H03M1/12 ,  H03M7/30 ,  H04B14/04 ,  H04Q1/457 ,  G10L9/14

公开(公告)号：CA2120558A1

公开(公告)日：1994-12-31

申请号：CA2120558

申请日：1994-04-05

Applicant: IBM

Inventor： GALAND CLAUDE ,  LEBIZAY GERALD ,  MAUDUIT DANIEL ,  MUNIER JEAN-MARIE ,  PAUPORTE ANDRE ,  SAINT-GEORGES ERIC ,  SPAGNOL VICTOR

IPC: H04L29/06 ,  H04L12/56

Abstract: A high performance data packet buffering method and a programmable data communication adapter for high speed packet transmission networks are disclosed. The line adapter includes programmable processing means, for receiving and transmitting data packets of fixed or variable length. This system is characterized in that it comprises means for buffering (132) said data packets, means for identifying said buffering means and said data packets in said buffering means, means for queueing (Figure 15) in storing means (131) said identifying means in a single instruction, means for dequeueing (Figure 16) from said storing (131) means said identifying means in another single instruction , means for releasing said buffering means, Each instruction comprises up to three operations executed in parallel by said processing means : an arithmetical and logical (ALU) operation on said identifying means, a memory operation on said storing means, and a sequence operation.

公开(公告)号：DE3888830T2

公开(公告)日：1994-11-24

申请号：DE3888830

申请日：1988-08-30

Applicant: IBM

Inventor： BELLOC JACQUES ,  GALAND CLAUDE ,  LANCON EMMANUEL ,  MILEWSKI ANDRE ,  QUINTIN MICHEL

IPC: H03H17/00 ,  H03H17/02 ,  H03H17/06

公开(公告)号：DE68914147D1

公开(公告)日：1994-04-28

申请号：DE68914147

申请日：1989-06-07

Applicant: IBM

Inventor： GALAND CLAUDE ,  MENEZ JEAN

IPC: G10L13/00 ,  G10L19/04 ,  G10L19/08 ,  G10L9/14

公开(公告)号：DE3688980T2

公开(公告)日：1994-04-21

申请号：DE3688980

申请日：1986-10-30

Applicant: IBM

Inventor： ROSSO MICHELE ,  GALAND CLAUDE

IPC: G10L19/02 ,  G10L19/00 ,  H03M1/12 ,  H03M7/30 ,  H04B1/66 ,  H04B14/04

Abstract: A process for multirate subband encoding voice signals. At least a portion of the original signal bandwidth is plit into p sub-bands the contents of which is to be coded using dynamic allocation of quantizing levels throughout the sub-bands. Let's assume the signal is to be coded for a set of coding rates R(1) Then the sub-band samples S(i,j) are quantized according to 1(i; R(K). And, finally the obtained quantized sub-band samples are embedded into a multirate frame.

公开(公告)号：DE3871369D1

公开(公告)日：1992-06-25

申请号：DE3871369

申请日：1988-03-08

Applicant: IBM

Inventor： BOTTAU FRANCOISE ,  GALAND CLAUDE ,  MENEZ JEAN ,  ROSSO MICHELE

IPC: G10L19/06 ,  G10L19/12 ,  H03M7/30 ,  G10L9/14

Abstract: The encoding process comprises the steps of filtering voice signal using a short-term predictive filter tuned by sets of partial correlation related coefficients derived from s(n) and deriving from them a short-term residual r(n). A long-term residual e(n) is derived by subtracting a weighted and delayed synthesis short-term residual from the r(n). E(n) is split into consecutive blocks. Each block of the long term residual e(n) is code excited encoded into one TABLE address reference k and one gain G.

公开(公告)号：DE3585034D1

公开(公告)日：1992-02-06

申请号：DE3585034

申请日：1985-10-30

Applicant: IBM

Inventor： PLATEL GUY ,  GALAND CLAUDE

IPC: H04B3/23

Abstract: The process includes measuring and delaying the energy of blocks of consecutive output samples to derive a delayed sequence of blocks of output energy values w(n). The energy of blocks of consecutive input samples is measured to derive blocks of input energy values v(n). The w(n) and v(n) are inter-correlated over fixed length periods. The block energies inter-correlation peak value is detected to determine the flat delay length. The length of the echo replica flat delay line is set accordingly. The flat delay is further adjusted through intercorrelating one block of delayed outgoing samples with one block of incoming samples. The block samples intercorrelation peak is detected to determine an accurate flat delay value and the length of the echo replica is further adjusted.

公开(公告)号：CA1227295A

公开(公告)日：1987-09-22

申请号：CA479412

申请日：1985-04-17

Applicant: IBM

Inventor： CALLENS PAUL ,  GALAND CLAUDE

IPC: H04M3/00 ,  H04J3/17 ,  H04L12/64

Abstract: A method of transmitting digitized voice signals as packets of bits in a digital transmission network to which a plurality of terminals acting as sources of voice signals are attached. The method includes, at the transmitting end and for each source, the steps of : dividing the signal into successive, digitally-coded segments; performing a so-called source activity detection; and assembling the bits resulting from the coding operation into packets, each packet being provided with a header containing, in particular, a parity bit whose logic value is modified whenever an inactivity or long pause is detected. At the receiving end, detection of a change in the value of said parity bit causes the delivery of a packet of bits to the decoder/synthesizer to be delayed by a given time interval called retention delay.

公开(公告)号：DE69927252T2

公开(公告)日：2006-06-29

申请号：DE69927252

申请日：1999-07-13

Applicant: IBM

Inventor： FICHOU ALINE ,  LE PENNEC JEAN-FRANCOIS ,  GALAND CLAUDE

IPC: H04L12/24 ,  H04L12/26 ,  H04L12/70 ,  H04Q11/04

Abstract: Disclosed herein is a method and a system of network capacity planning for use in a high speed packet switching network. The network comprises a plurality of switching nodes interconnected through a plurality of communication links, each of the switching nodes comprises means for switching packets from at least one input link to at least one output link. Each of the output links is coupled to at least one buffer in the switching node for queuing packets before they are transmitted over the output link. The network capacity planning method of the invention comprises the following steps: In each of the switching nodes and for each of the output links, a time distribution of the occupancy of each buffer during a predetermined monitoring time period is measured, and stored in a centralized memory location. Then, the buffer occupancy time distribution data are retrieved from the centralized memory location in all the switching nodes, gathered and stored in a network dedicated server. These buffer occupancy time distribution data for all the nodes are transmitted from the dedicated network server to a network monitoring center. In the network monitoring center, the buffer occupancy time distribution data are integrated on a larger time scale, and stored. Finally, the integrated buffer occupancy time distribution data are used to apply network capacity planning actions to the network resources.